whatsup party people

hey everyone

my name's adrian and i'm very eager to start turking. i accidentally mistyped my social for amazon payments so i'm getting it all taken care of before i can get going. i plan on really going all out for hours and hours of just non stop focus and speed.

until then i've been researching my a** off trying to learn as much as i can from forums, reddit, fb groups, even some discords. the mturk community is definitely very helpful.

from what i've gathered i'd like to eventually specialize in batches that can be scripted because they seem like the most lucrative opportunity. when i'm starting out though, i'm just going to get a feel for the marketplace with low rejection hits, getting my numbers up, and qualifications.

i'm going to be using mturk suite and panda crazy but i've seen people recommend overwatch (this thread here). from my understanding, the only difference between ow and pc is pc pandas requesters and ow can panda requesters and specific hits but i feel like there's something i'm missing.

i'll be sure to start checking out the daily threads, i see they're all very active. if anyone is a newbie like me and has any questions, feel free to ask because i've been doing a lot of reading and i see the same questions over and over.

looking forward to connecting with everyone!

 

Welcome! We're excited to have ya!

 

Welcome to the Hub!

Looks like you're getting yourself set up pretty well with your research already. That's great! I think you have PandaCrazy and Overwatch uses a little backwards though. PandaCrazy is used best to panda individual hits/batches. While it can search for requesters I wouldn't recommend using up a page request for that. Overwatch searches for the requesters you have on the list to notify you if they've posted anything. Overwatch only uses one request for your entire requester list.

 

i see so i had the two backwards. so pc is best used for hits, ow is best used for requesters.

right now i'm thinking a lot about my "work flow" or how i'm going to use scripts to find hits. is this on track:

1. run two instances of hit finder, one for surveys and batches
2. use panda crazy when i find a batch that's getting low on hits available (do i panda a hit once and then delete it since they come and go?)
3. as i find good requesters, add them to my overwatch

i want to optimize how much work i can get done on mturk and i feel like the key is to come up with a system that just continuously finds hits so that's what i'm working towards. but i know my approach may be a little off because my inexperience.

 

You'll probably want to do some trial and error and see what works best for you. Personally, I just run one Hit Finder and use the block list a lot to filter out the stuff I don't want to see clogging up the list. I use it for both batches and surveys. That way it's only one page request and I get to see both.

As for Pandas on PC, it depends. Some batches come back every so often and some come back regularly. You'll start to notice which ones by watching the daily thread. Keep the ones you like to work on that come back. You can make tabs in PC and put different sets of hits in the different tabs to not clog up the front page.

And yes, as you find requesters you like to do work for, add the to overwatch.

 

@Celtic Lady thanks that clears up a lot. i wasn't sure if hits come and go or if they could be "reloaded". thanks for taking the time

i read in this thread that people using hit finder will set the timer to 5 sec so it won't miss anything and that it can save a log of hits it might have missed. do you know where the 5 seconds comes from? also, what exactly is this saved log because i'm having a hard time understanding exactly what it does?

 

This again will come down to you testing it out some and seeing how it works best for you. I use a 1 second interval and have it scrape the worker.mturk site rather than www.mturk. They each have different page request pools. If you are requesting pages too often (scraping, pandas, submitting hits, updating hit database mkII etc) you will get a PRE. <--click where it's blue to read about them. So, I run it at 1 sec on worker and work on hits + panda on www. That's just the way I've been doing it. It may not be optimal and I've been experimenting with running different things between the two mturk sites and haven't completely settled on what works best for me yet.

The logged hits show up in the bottom portion of the window below the current found hits section. The hits in the logged section may or may not be currently active hits but you can still get the link to add a panda if you want to try and catch a return/expire/new one.


They are probably using 5 seconds so it doesn't use a page request so often and they are still able to see nearly all hits show up at least briefly and then they are in the logged section for pandas. I prefer 1 second or 2 because hits go very, very quickly. I do both surveys and batches and especially high paying surveys are gone before you can click on them even at a 1 sec scrape a lot of the time.

 

thanks x1000 from all the reading i've done i got a fair idea of how things work but your replies really filled in the blanks for me. all your info will be put to good use. really appreciate you for taking the time

 

Welcome to the Hub, homie. Seconding everyone who commended you on doing your homework. You'll fit in well and be a top earner in no time. Turking can be great if you know your stuff and have the motivation to find what works best for you. See you around the forum

 

This headline sounds like Chip Chipperson... >_>